Nof S.Y. Springer Handbook of Automation

Подождите немного. Документ загружается.

Teleoperation References 465

References

27.1 T. Sheridan: Telerobotics, Automation and Human

Supervisory Control (MIT Press, Cambridge 1992)

27.2 E. Nu

no,A.Rodríguez,L.Basa

nez: Force reﬂect-

ing teleoperation via IPv6 protocol with geometric

constraints haptic guidance. In: Advances in Teler-

obotics, STAR, Vol. 31 (Springer, Berlin, Heidelberg

2007) pp. 445–458

27.3 R.J. Anderson, M.W. Spong: Bilateral control of

teleoperators with time delay, IEEE Trans. Autom.

Control 34(5), 494–501 (1989)

27.4 G. Niemeyer, J.J.E. Slotine: Stable adaptive teleop-

eration, IEEE J. Ocean. Eng. 16(1), 152–162 (1991)

27.5 P. Arcara, C. Melchiorri: Control schemes for tele-

operation with time delay: A comparative study,

Robot. Auton. Syst. 38,49–64(2002)

27.6 P.F. Hokayem, M.W. Spong: Bilateral teleoperation:

An historical survey, Automatica 42, 2035–2057

(2006)

27.7 R. Ortega, N. Chopra, M.W. Spong: A new pas-

sivity formulation for bilateral teleoperation with

time delays, Proc. CNRS-NSF Workshop: Advances

in time-delay systems (Paris, 2003)

27.8 E. Nu

no,L.Basa

nez, R. Ortega: Passive bilateral

teleoperation framework for assisted robotic tasks,

Proc. IEEE Int. Conf. Robot. Autom. (Rome, 2007)

pp. 1645–1650

27.9 S. Munir, W.J. Book: Control techniques and pro-

gramming issues for time delayed internet based

teleoperation, ASME J. Dyn. Syst. Meas. Control

125(2), 205–214 (2004)

27.10 S.E. Salcudean, M. Zhu, W.-H. Zhu, K. Hashtrudi-

Zaad: Transparent bilateral teleoperation under

position and rate control, Int. J. Robot. Res. 19(12),

1185–1202 (2000)

27.11 N. Chopra, M.W. Spong, R. Ortega, N. Barbanov:

On tracking performance in bilateral teleoperation,

IEEE Trans. Robot. 22(4), 844–847 (2006)

27.12 T. Namerikawa, H. Kawada: Symmetric impedance

matched teleoperation with position tracking,

Proc. 45th IEEE Conf. Decis. Control (San Diego,

2006) pp. 4496–4501

27.13 E. Nu

no, R. Ortega, N. Barabanov, L. Basa

nez: A

globally stable proportional plus derivative con-

troller for bilateral teleoperators, IEEE Trans. Robot.

24(3), 753–758 (2008)

27.14 R. Lozano, N. Chopra, M.W. Spong: Convergence

analysis of bilateral teleoperation with constant

human input, Proc. Am. Control Conf. (New York

2007) pp. 1443–1448

27.15 R. Lozano, N. Chopra, M.W. Spong: Passivation of

force reﬂecting bilateral teleoperators with time

varying delay, Proc. Mechatron. Conf. (Entschede

2002)

27.16 N. Chopra, M.W. Spong: Adaptive synchroniza-

tion of bilateral teleoperators with time delay. In:

Advances in Telerobotics, STAR, Vol. 31 (Springer,

Berlin, Heidelberg 2007) pp. 257–270

27.17 C. Secchi, S. Stramigioli, C. Fantuzzi: Variable de-

lay in scaled port-Hamiltonian telemanipulation,

Proc. 8th Int. IFAC Symp. Robot Control (Bologna

2006)

27.18 M. Boukhnifer, A. Ferreira: Wave-based passive

control for transparent micro-teleoperation sys-

tem, Robot. Auton. Syst. 54(7), 601–615 (2006)

27.19 P. Loshin: IPv6, Theory, Protocol, and Practice,2nd

edn. (Morgan Kaufmann, San Francisco 2003)

27.20 W.R. Sherman, A. Craig: Understanding Virtual Re-

ality. Interface, Application and Design (Morgan

Kaufmann, San Francisco 2003)

27.21 R. Azuma: A survey of augmented reality, Presence

Teleoper. Virtual Environ. 6(4), 355–385 (1997)

27.22 M. Inami, N. kawakami, S. Tachi: Optical camou-

ﬂage using retro-reﬂective projection technology,

Proc. Int. Symp. Mixed Augment. Real. (Tokyo 2003)

pp. 18–22

27.23 R. Azuma, Y. Baillot, R. Behringer, S. Feiner,

S. Julien, B. MacIntyre: Recent advances in aug-

mented reality, IEEE Comp. Graphics Appl. 21(6),

34–47 (2001)

27.24 A. Rastogi, P. Milgram, J. Grodski: Augmented

telerobotic control: A visual interface for un-

structured environments, Proc. KBS/Robot. Conf.

(Montreal 1995)

27.25 A.Kron,G.Schimdt,B.Petzold,M.Zäh,P.Hinter-

seer, E. Steinbach: Disposal of explosive ordnances

by use of a bimanual haptic system, Proc. IEEE Int.

Conf. Robot. Autom. (New Orleans 2004) pp. 1968–

1973

27.26 A. Ansar, D. Rodrigues, J. Desai, K. Daniilidis,

V. Kumar, M. Campos: Visual and haptic collabora-

tive tele-presence, Comput. Graph. 25(5), 789–798

(2001)

27.27 B. Dejong, E. Faulring, E. Colgate, M. Peshkin,

H. Kang, Y. Park, T. Erwing: Lessons learned froma

novel teleoperation testbed, Ind. Robot: Int. J.

33(3), 187–193 (2006)

27.28 Y. Xiong, S. Li, M. Xie: Predictive display and inter-

action of telerobots based on augmented reality,

Robotica 24, 447–453 (2006)

27.29 S. Otmane, M. Mallem, A. Kheddar, F. Chavand: Ac-

tive virtual guides as an apparatus for augmented

reality based telemanipulation system on the in-

ternet, Proc. 33rd Annu. Simul. Symp. (Washington

2000) pp. 185–191

27.30 J.Gu,E.Auguirre,P.Cohen:Anaugmentedre-

ality interface for telerobotic applications, Proc.

Workshop Appl. Comput. Vis. (Orlando 2002)

pp. 220–224

27.31 A. Rodríguez, L. Basa

nez, E. Celaya: A Relational

Positioning Methodology for Robot Task Speciﬁ-

Part C 27

466 Part C Automation Design: Theory, Elements, and Methods

cation and Execution, IEEE Trans. Robot. 24(3),

600–611 (2008)

27.32 Y.H. Liu: Computing n-ﬁnger form-closure grasps

on polygonal objects, Int. J. Robot. Res. 19(2), 149–

158 (2000)

27.33 D. Ding, Y. Liu, S. Wang: Computation of 3-D form-

closure grasps, IEEE Trans. Robot. Autom. 17(4),

515–522 (2001)

27.34 M. Roa, R. Suárez: Finding locally optimum force-

closure grasps, Robot. Comput.-Integr. Manuf. 25,

536–544 (2009)

27.35 J. Cornell

a, R. Suárez, R. Carloni, C. Melchiorri: Dual

programming based approach for optimal grasp-

ing force distribution, Mechatronics 18(7), 348–356

(2008)

27.36 T.B.Martin,R.O.Ambrose,M.A.Diftler,R.Platt,

M.J. Butzer: Tactile gloves for autonomous grasping

with the NASA/DARPA Robonaut, Proc. IEEE Int. Conf.

Robot. Autom. (New Orleans 2004) pp. 1713–1718

27.37 M.Bergamasco,A.Frisoli,C.A.Avizzano:Exo-

skeletons as man–machine interface systems for

teleoperation and interaction in virtual environ-

ments. In: Advances in Telerobotics STAR Ser., Vol.

31 (Springer, New York 2007) pp. 61–76

27.38 S.B. Kang, K. Ikeuchi: Grasp recognition using the

contact web, Proc. IEEE/RSJ Int. Conf. Intell. Robots

Syst. (Raleigh 1992) pp. 194–201

27.39 R.L. Feller, C.K.L. Lau, C.R. Wagner, D.P. Pemn,

R.D. Howe: The effect of force feedback on remote

palpation, Proc. IEEE Int. Conf. Robot. Autom. (New

Orleans 2004) pp. 782–788

27.40 M. Benali-Khoudja, M. Hafez, J.M. Alexandre,

A. Kheddar: Tactile interfaces: a state-of-the-art

survey, Proc. 35th Int. Symp. Robot. (Paris 2004)

pp. 721–726

27.41 T. Wotjara, K. Nonami: Hand posture detection by

neural network and grasp mapping for a master–

slave hand system, Proc. IEEE/RSJ Int. Conf. Intell.

Robots Syst. (Sendai 2004) pp. 866–871

27.42 M. Roa, R. Suárez: Independent contact regions for

frictional grasps on 3-D objects, Proc. IEEE Int. Conf.

Robot. Autom. (Pasadena 2008)

27.43 K.B. Shimoga: Robot grasp synthesis algorithms: a

survey,Int.J.Robot.Res.15(3), 230–266 (1996)

27.44 R. Suárez, P. Grosch: Mechanical hand MA-I as ex-

perimental system for grasping and manipulation,

Video Proc. IEEE Int. Conf. Robot. Autom. (Barcelona

2005)

27.45 A. Iborra, J.A. Pastor, B. Alvarez, C. Fernandez,

J.M. Fernandez: Robotics in radioactive environ-

ments, IEEE Robot. Autom. Mag. 10(4), 12–22 (2003)

27.46 W. Book, L. Love: Teleoperation telerobotics tele-

presence. In: Handbook of Industrial Robotics,2nd

edn. (Wiley, New York 1999) pp. 167–186

27.47 R. Aracil, M. Ferre: Telerobotics for aerial live power

line maintenance. In: Advances in Telerobotics,

STAR Ser., Vol. 31, (Springer, Berlin, Heidelberg 2007)

459–469

27.48 J.H.Dunlap,J.M.VanName,J.A.Henkener:Robotic

maintenance of overhead transmission lines, IEEE

Trans. Power Deliv. 1

(3), 280–284 (1986)

27.49 R.Aracil,M.Ferre,M.Hernando,E.Pinto,J.M.Se-

bastian: Telerobotic system for live-power line

maintenance: ROBTET, Control Eng. Prac. 10(11),

1271–1281 (2002)

27.50 C.T. Haas, Y.S. Ki: Automation in infrastructure con-

struction, Constr. Innov. 2,191–210(2002)

27.51 Y. Hiramatsu, T. Aono, M. Nishio: Disaster restora-

tion work for the eruption of Mt Usuzan using an

unmanned construction system, Adv. Robot. 16(6),

505–508 (2002)

27.52 A.M. Lytle, K.S. Saidi, R.V. Bostelman, W.C. Stones,

N.A. Scott: Adapting a teleoperated device for

autonomous control using three-dimensional po-

sitioning sensors: experiences with the NIST

RoboCrane, Autom. Constr. 13, 101–118 (2004)

27.53 A.J. Kwitowski, W.D. Mayercheck, A.L. Brautigam:

Teleoperation for continuous miners and haulage

equipment, IEEE Trans. Ind. Appl. 28(5), 1118–1125

(1992)

27.54 G. Baiden, M. Scoble, S. Flewelling: Robotic sys-

tems development for mining automation, Bull.

Can.Inst.Min.Metall.86.972, 75–77 (1993)

27.55 J.C. Ralston, D.W. Hainsworth, D.C. Reid, D.L. An-

derson, R.J. McPhee: Recent advances in remote

coal mining machine sensing, guidance, and tele-

operation, Robotica 19(4), 513–526 (2001)

27.56 A.J. Park, R.N. Kazman: Augmented reality for min-

ing teleoperation, Proc. of SPIE Int. Symp. Intell.

Syst.Adv.Manuf.–Telemanip.TelepresenceTech-

nol. (1995) pp. 119–129

27.57 T.J. Nelson, M.R. Olson: Long delay telecontrol of

lunar mining equipment, Proc. 6th Int. Conf. Expo.

Eng., Constr., Oper. Space, ed. by R.G. Galloway,

S.Lokaj(Am.Soc.Civ.Eng.,Reston1998)pp.477–

484

27.58 N. Wilkinson: Cooperative control in teleoperated

mining environments, 55th Int. Astronaut. Congr.

Int. Astronaut. Fed., Int. Acad. Astronaut. Int. Inst.

Space Law (Vancouver 2004)

27.59 P. Ridao, M. Carreras, E. Hernandez, N. Palom-

eras: Underwater telerobotics for collaborative

research. In: Advances in Telerobotics,STARSer.,

Vol. 31 (Springer, Berlin, Heidelberg 2007) pp. 347–

359

27.60 S. Harris, R. Ballard: ARGO: Capabilities for deep

ocean exploration, Oceans 18,6–8(1986)

27.61 M. Fontolan: Prestige oil recovery from the sunken

part of the Wreck, PAJ Oil Spill Symp. (Petroleum

Association of Japan, Tokyo 2005)

27.62 G. Antonelli (Ed.): Underwater Robots: Motion

and Force Control of Vehicle–Manipulator Systems

(Springer, Berlin 2003)

27.63 C. Canudas-de-Wit, E.O. Diaz, M. Perrier: Ro-

bust nonlinear control of an underwater vehi-

cle/manipulator system with composite dynamics,

Part C 27

Teleoperation References 467

Proc. IEEE Int. Conf. Robot. Autom. (Leuven 1998)

pp. 452–457

27.64 M. Lee, H-S. Choi: A robust neural controller for

underwater robot manipulators, IEEE Trans. Neural

Netw. 11(6), 1465–1470 (2000)

27.65 J. Kumagai: Swimming to Europe, IEEE Spectrum

44(9), 33–40 (2007)

27.66 L. Pedersen, D. Kortenkamp, D. Wettergreen,

I. Nourbakhsh: A survey of space robotics, Proc.

7th Int. Symp. Artif. Intell., Robot. Autom. Space

(Nara 2003)

27.67 F. Doctor, A Glas, Z. Pronk: Mission prepara-

tion support of the European Robotic Arm (ERA),

National Aerospace Laboratory report NLR-TP-

2002-650 (Netherlands 2003)

27.68 S.Roderick,B.Roberts,E.Atkins,D.Akin:The

ranger robotic satellite servicer and its autonomous

software-based safety system, IEEE Intell. Syst.

19(5), 12–19 (2004)

27.69 W. Bluethmann, R. Ambrose, M. Diftler, S. Askew,

E. Huber, M. Goza, F. Rehnmark, C. Lovchik, D. Ma-

gruder: Robonaut: A robot designed to work with

humans in spaces, Auton. Robots 14, 179–197 (2003)

27.70 G. Hirzinger, B. Brunner, K. Landzettel, N. Sporer,

J. Butterfass, M. Schedl: Space Robotics – DLR’s

telerobotic concepts, lightweight arms and articu-

lated hands, Auton. Robots 14, 127–145 (2003)

27.71 S.E. Fredrickson, S. Duran, J.D. Mitchell: Mini AER-

Cam inspection robot for human space missions,

AIAA Space 2004 Conf. Exhib. (San Diego 2004)

27.72 T. Imaida, Y. Yokokohji, T. Doi, M. Oda,

T. Yoshikawa: Ground–space bilateral teleoper-

ation of ETS-VII robot arm by direct bilateral

coupling under 7-s time delay condition, IEEE

Trans. Robot. Autom. 20(3), 499–511 (2004)

27.73 R.A. Lindemann, D.B. Bickler, B.D. Harrington,

G.M. Ortiz, C.J. Voorhees: Mars exploration rover

mobility development, IEEE Robot. Autom. Mag.

13(2), 19–26 (2006)

27.74 G.A. Landis: Robots and humans: synergy in plan-

etary exploration, Acta Astronaut. 55(12), 985–990

(2004)

27.75 J.W. Hill, P.S. Green, J.F. Jensen, Y. Gorfu,

A.S. Shah: Telepresence surgery demonstration

system, Proc. IEEE Int. Conf. Robot. Autom. (IEEE

Computer Society, San Diego 1994) pp. 2302–2307

27.76 J. Marescaux, J. Leroy, M. Gagner, F. Rubino,

D. Mutter, M. Vix, S.E. Butner, M. Smith: Transat-

lantic robot-assisted telesurgery, Nature 413,

379–380 (2001)

27.77 S.E. Butner, M. Ghodoussi: A real-time system for

tele-surgery, Proc. 21st Int. Conf. Distrib. Com-

put. Syst. (IEEE Computer Society, Washington 2001)

pp. 236–243

27.78 G.S. Guthart, J.K. Jr. Salisbury: The Intuitive

telesurgery system: overview and application,

Proc. IEEE Int. Conf. Robot. Autom. (San Francisco

2000) pp. 618–621

27.79 M. Li, A. Kapoor, R.H. Taylor: Telerobotic control

by virtual ﬁxtures for surgical applications. In:

Advances in Telerobotics, STAR, Vol. 31 (Springer,

Berlin, Heidelberg 2007) pp. 381–401

27.80 A. Smith, J. Smith, D.G. Jayne: Telerobotics: surgery

for the 21st century, Surgery 24(2), 74–78 (2006)

27.81 M. Topping: An Overview of the development of

Handy 1, a rehabilitation robot to assist the severely

disabled, J. Intell. Robot. Syst. 34(3), 253–263

(2002)

27.82 C. Balaguer, A. Giménez, A. Jardón, R. Correal,

S. Martínez, A.M. Sabatini, V. Genovese: Proprio,

teleoperation of a robotic system for disabled per-

sons’ assistance in domestic environments. In:

Advances in Telerobotics, STAR, Vol. 31 (Springer,

Berlin, Heidelberg 2007) pp. 415–427

27.83 O. Reinoso, C. Fernández, R.

Neco: User voice as-

sistance tool for teleoperation. In: Advances in

Telerobotics, STAR Ser., Vol. 31 (Springer, Berlin,

Heidelberg 2007) pp. 107–120

27.84 K. Kawamura, M. Iskarous: Trends in service

robots for the disabled and the elderly, Proc.

IEEE/RSJ/GI Int. Conf. Intell. Robots Syst. (Munich

1994) pp. 1647–1654

27.85 G. Pires, U. Nunes: A wheelchair steered through

voice commands and assisted by a reactive fuzzy

logic controller, J. Intell. Robot. Syst. 34(3), 301–314

(2002)

27.86 P. Hoppenot, E. Colle: Human-like behavior robot-

application to disabled people assistance, Proc.

IEEE Int. Conf. Syst., Man, Cybern. (Soc. Syst. Man

Cyber., Nashville 2000) pp. 155–160, .

27.87 M.E.Pollack,S.Engberg,J.T.Matthews,S.Thrun,

L. Brown, D. Colbry, C. Orosz, B. Peintner,

S. Ramakrishnan, J. Dunbar-Jacob, C. McCarthy,

M. Montemerlo, J. Pineau, N. Roy: Pearl: A mo-

bile robotic assistant for the elderly, AAAI Workshop

Autom. Eldercare (Alberta 2002)

27.88 T. Wojtara, K. Nonami, H. Shao, R. Yuasa, S. Amano,

D. Waterman, Y. Nobumoto: Hydraulic master–

slave land mine clearance robot hand controlled

by pulse modulation, Mechatronics 15,589–609

(2005)

27.89 K. Kato, S. Hirose: Development of the quadruped

walking robot, TITAN-IX–mechanical design con-

cept and application for the humanitarian demi-

ning robot, Adv. Robot. 15(2), 191–204 (2001)

27.90 P. Gonzalez de Santos, E. Garcia, J.A. Cobano,

A. Ramirez: SILO6: A six-legged robot for humani-

tarian demining tasks, Proc. 10th Int. Symp. Robot.

Appl. World Autom. Congr. (2004)

27.91 J.-D. Nicoud: Vehicles and robots for humanitarian

demining, Ind. Robot 24(2), 164–168 (1997)

27.92 M.K. Habib: Humanitarian demining: reality and

the challenge of technology – the state of the arts,

Int. J. Adv. Robot. Syst. 4(2), 151–172 (2007)

27.93 C.S. Tzafestas, N. Palaiologou, M. Alifragis: Vir-

tual and remote robotic laboratory: comparative

Part C 27

468 Part C Automation Design: Theory, Elements, and Methods

experimental evaluation, IEEE Trans. Educ. 49(3),

360–369 (2006)

27.94 X.Giralt,D.Jofre,R.Costa,L.Basa

nez: Proyecto

de Laboratorio Remoto de Automática: Objetivos

y Arquitectura Propuesta, III Jornadas de Trabajo

EIWISA 02, Ense

nanza vía Internet/Web de la In-

geniería de Sistemas y Automática (Alicante, 2002)

pp. 93–98, in Spanish

27.95 M. Alencastre, L. Munoz, I. Rudomon: Teleoperat-

ing robots in multiuser virtual environments, Proc.

4th Mexican Int. Conf. Comp. Sci. (Tlaxcala 2003)

pp. 314–321

Part C 27

469

Distributed A

28. Distributed Agent Software for Automation

Francisco P. Maturana, Dan L. Carnahan, Kenwood H. Hall

Agent-based software and hardware technologies

have emerged as a major approach to organize

and integrate distributed elements of complex au-

tomation. As an example, this chapter focuses on

a particular situation. Composite curing, a rapidly

developing industry process, generates high costs

when not properly controlled. Curing autoclaves

require tight control over temperature that must

be uniform throughout the curing vessel. This

chapter discusses how agent-based software is

being implemented into the curing process by

pushing control logic down to the lowest level

of the control hierarchy into the process con-

troller, i. e., the programmable logic controller

(PLC). The chapter also discusses how the ben-

eﬁts of process survivability, diagnostics, and

dynamic reconﬁguration are achieved through the

use of autoclave and thermocouple intelligent

agents.

28.1 Composite Curing Background ............... 471

28.2 Industrial Agent Architecture................. 473

28.2.1 Agent Design

and Partitioning Perspective........ 474

28.2.2 Agent Tool................................. 475

28.3 Building Agents for the Curing System.... 475

28.4 Autoclave and Thermocouple Agents ...... 477

28.4.1 Autoclave Agent ......................... 477

28.4.2 Thermocouple Agent................... 478

28.5 Agent-Based Simulation ....................... 478

28.6 Composite Curing Results

and Recommendations ......................... 480

28.6.1 Designing the Validation System .. 480

28.6.2 Modeling Process Dynamics ......... 481

28.6.3 Timing and Stability Criteria ........ 484

28.7 Conclusions .......................................... 484

28.8 Further Reading ................................... 484

References .................................................. 485

A general introduction and overview of agent-based

automation can be found in the additional reading

listed. Autoclave curing is vital to the production

and support of many industries. In fact, there are

so many industries that are dependent upon auto-

clave curing that improvements to curing production

controls would lend a competitive advantage to the

corporation that could deliver these improvements.

While the focus of this particular work is directed

predominately towards the production of compos-

ite materials, there are many other processes that

could beneﬁt from this work, with almost no dif-

ferences in terms of the autoclave curing controls.

This is especially true for the following products

that require autoclave curing: composite materials,

polymers, rubber, concrete products (e.g., sand-lime

brick, asbestos, hydrous calcium silicate, cement (steam

curing)), tobacco, textiles, electrical and electronics

products (e.g., printed circuit boards (PCBs), ceramic

substrates) chemical, medical/pharmaceutical products,

wood and buildingproducts, metallurgical products,tire

rethreading, glass laminating, and aircraft/aerospace

products.

Much work has recently been reported for new cur-

ing methods and autoclave curing devices. This is one

indication that there is a trend towards researching and

developing new and better ways of curing materials.

This research and development (R&D) has increased

to meet market demands and technical papers regard-

ing materials curing is proliferating. Market research

studies also support the notion that there is increased

demand for composite materials and that this trend

will continue in the future. With increased demand

comes a need for increased efﬁciencies in the curing

production process. Composite material manufacturers,

aerospace manufacturers, and manufacturers of other

Part C 28

470 Part C Automation Design: Theory, Elements, and Methods

materials that require curing are increasing their capital

expenditure budgets for the development and/or acqui-

sition of improved autoclave manufacturing capabilities

to meet this demand head-on.

Composite curing is accomplished through the

proper application of heating and cooling to composite

material inside autoclaves or automated ovens. De-

pending upon the type of composite material or the

application of the composite material, a different set of

curing parameters is used to control the curing process.

Composite materials are cured under very stringent

speciﬁcations, especially for materials that are used for

aerospace applications. Ifcomposites are curedat atem-

perature that is too high, the material could become

brittle and will be susceptible to breaking. If cured at

a temperature that is too low, the material may not bond

correctly and will eventually come apart.

The speciﬁcations that govern the curing process

for composite materials are called recipes or proﬁles.

Recipes/proﬁles differ slightly for an autoclave that has

convection heaters versus an autoclave that has gas

heaters. Likewise, an autoclave that has cryogenic cool-

ers versus an autoclave that is air-cooled or fan-cooled

would have slightly different control parameters within

its control recipe. In the case of autoclave operations

there are, generally, three proﬁles used to control the

composite curing process: temperature, pressure, and

vacuum. Of these three, temperature is the most im-

portant and the most difﬁcult to control. As seen in

Fig.28.1, six separate periods control the temperature

within the curing autoclave. Temperature increases and

decreases are ramped up or down according to a partic-

ular rate of change (i.e., a slope) that is also regulated

within the speciﬁcation.

0 27 53 79 105 131 157 183 209 235 261 287

Temperature profile

Temperature (°C)

Time (min)

400

350

300

250

200

150

100

Fig. 28.1 Typical autoclave temperature proﬁle

State-of-the-art systems use centralized material

management control loops to guide the curing activ-

ity through a prescribed set of thermal, pressure, and

vacuum proﬁles. Controlling autoclave operations so

that they conform to thermal proﬁles is the main con-

cern of the system and, hence, dictates the requirements

for system control. For a composite material to bond

properly, temperature throughout an autoclave must be

maintained at a very speciﬁc level. Arrays of ther-

mocouples provide temperature readings back to the

control program through output data cards and are used

to controlthe operationof heaters and, sometimes, cool-

ers. One of the challenges of controlling temperature

within theautoclave chamber is tokeep it within a range

of temperature values even when the temperature must

be modiﬁed several times during the production pro-

cess (as shown in Fig.28.1). The nonlinear behaviors

of a temperature proﬁle are monitored by a lead ther-

mocouple that is designated based upon its position in

the autoclave, the type of material (or materials) be-

ing cured, chamber idiosyncrasies, and other production

variables.

Whichever thermocouple is selectedas the lead sen-

sor, the feedback temperature data from that sensor will

not be allowed to exceed the temperature proﬁle tem-

perature during heating operations. During the period

of time when the thermal proﬁle speciﬁes a decrease

in temperature (i.e., cooling operations), the lead ther-

mocouple is usually used to determine the minimum

temperature that will be allowed to occur. Naturally,

for this type of control to be effective, network con-

nectivity is critical to maintain high-level material state

control. If a lead thermocouple disconnects from the

composite material while the composite is being cured,

the disconnection must be detected as soon as possi-

ble, and another thermocouple must be selected by the

software to act as the lead. The former lead thermocou-

ple is ignored from that point forward while the curing

process continues. For cooling operations, the thermo-

couple that is used to control the cooling is actually

called the lag thermocouple and temperatures are not

allowed to go below that of the lag sensor.

On the surface of thiscontrol application, it doesnot

seem that the requirements would be difﬁcult to accom-

plish through ordinary ladder logic routines. However,

there is a commercial issue that takes precedence over

the control issue. Millions of dollars of composite ma-

terials could be lost during one imperfect curing process

run. The survivability of the curing session must be en-

sured with greater probability despite perturbations in

the production process.

Part C 28

Distributed Agent Software for Automation 28.1 Composite Curing Background 471

Classical control methods sufﬁce to monitor and

control the thermal process within the autoclaves.

Although the curing activity is performed in a con-

trolled environment, there are dynamic perturbations

that affect the thermocouples, which could generate

unsatisfactory results, provoking complete rejection of

an expensive piece of composite material. Perturba-

tions of interest relate to potential malfunctioning in

the thermocouple themselves. A malfunctioning ther-

mocouple may appear healthy on visual inspection

but its internal operations may be faulty. This is the

case when thermocouples generate bogus readings.

This type of problem is very difﬁcult to detect ofﬂine

and is not detected until the curing process has un-

dergone several steps. Classical control programs that

reside in the controller have limitations in terms of

enabling enough reasoning to detect such problems

early.

Another type of problem occurs when thermo-

couples detach from the material during curing. The

autoclave is a sealed controlled environment that cannot

be easily interrupted to reattach the sensors. The con-

troller must be capable of reacting to the failing reader

by performing corrective actions on the ﬂy without dis-

rupting operation.

A viable solution to the problem above is to aug-

ment the reasoning capability of the control system

with more sophisticated reasoning algorithms. These

augmented algorithms follow the process to generate

a model from it. Condition monitoring rules can then

be added to detect malfunctioning sensors. Typically, in

industrial implementations, such advanced capabilities

are placed in the personal computer (PC)level.APC

workstation is added to supervise the control system.

This approach converts the solution into a centralized

system. A centralized system suffers from other prob-

lems such as single point of failure and connectivity

issues, which exacerbate the problem of maintaining

a robust system for the whole duration of the process.

Current trends in industry and industrial techniques

show a strong tendency to move away from centralized

supervisory models.

The endeavor in this work is to show a new tech-

nique to cope with the issues above. A programmable

logic controller (PLC) can be designed with agent

capabilities to enable for advanced reasoning at the

controller level. Agents are software components that

encapsulate physical equipment knowledge (rules) and

properties in the form of capabilities, behaviors, and

procedures [28.1–3]. The capabilities express the type

of functions that an agent contributes to the well-being

of the system. Each capability is a construct of behav-

iors. Moreover, each behavior is made of sequentially

organized procedures. In the example, the agents are

seamless integrated with the control algorithm inside

the PLC.

The fundamental technical milestone is to eliminate

the connectivity problem, and to augment the reconﬁg-

uration of the control system by moving the condition

monitoring capabilities closer to the physical process,

into the PLC. The PLC is the core of the control op-

eration and therefore these devices have been widely

adopted in industrial automation with the necessary re-

dundancy to prevent losses.

The intention is to show fundamental aspects of

the agent-based control technology to solve a particular

composite curing application.The technologydiscussed

is intended for general use; it is possible to use the same

infrastructure to solve different applications.

The discussion will begin with a little bit of back-

ground about curing technology. Then, industrial agent

technology will be discussed. The chapter concludes

with an example and discussion of results on how to

model the curing application with intelligent agents.

28.1 Composite Curing Background

There are several markets in which large growth in

demand for autoclave cured products and materials

is forecast. The carbon ﬁber composites market has

been experiencing a growth of 12% during the last

23years and is expected to grow to US$12.2 billion

by 2011 [28.4, 5]. Carbon ﬁber reinforced thermo-

plastics will be improved by 37% between 2006 and

2010 [28.6].The opportunities for continuous ﬁber rein-

forced thermoplastic composites are likewise increasing

at a strong rate of growth. During 2002 alone, the

growth rate was 93% [28.7,8]. The aircraft maker Air-

bus has projected increases of thermoplastic composite

use of 20% per year.

The natural rubber market is also seeing higher

demands for its products. Rubber products have the

second largest need for autoclave curing production

improvements. Higherdemands for natural rubber prod-

ucts are largely due to the spectacular economic growth

Part C 28.1

472 Part C Automation Design: Theory, Elements, and Methods

in the People’s Republic of China (PRC) and greater

demand for consumer as well as industrial products.

Production levels just within Thailand have risen from

1.805million tons in1995 to 2.990 million tons in 2005.

Actual production exceeded 1995 demand projections

by 6.25% in 2000 and 40.05% in 2005. By 2010, it is

expected that 3.148million tons of natural rubber will

have to be produced. This will be more than 63% than

was originally thought to be needed in 1995.

There have been many technical and scientiﬁc

papers published recently that address many differ-

ent methods of curing within an autoclave. In 2004,

Salagnac et al. sought to improve the curing pro-

cess through heat exchanges that were mainly free

convection and radiation within a narrow-diameter

autoclave [28.9, 10]. Transient thermal modeling of

in situ curing during the tape winding of compos-

ite cylinders was attempted and successfully executed

by Kim et al. at the University of Texas at Austin

in 2002 [28.11]. Chang et al. studied the optimal de-

sign of the cure cycle for the consolidation of thick

composite laminates in 1996 at the National Cheng-

Kung University in Taiwan [28.12]. Smart autoclave

processing of thermoset resin matrix composites that

were based upon temperature and internal strain mon-

itoring was studied by Jinno et al. in 2003 [28.13].

Their work looked at the control of temperature

ramp rate so that the peak temperature predicted by

Springer’s thermochemical model is kept below an al-

lowable value. Cure completion was determined by

a cure rate equation, and internal strain monitoring

with embedded optical-ﬁber sensors [28.13]. In 2003,

a report was performed by Mawardi and Pitchumani

at the University of Connecticut [28.14]. Their work

was depicted as the development of optimal temper-

ature and current cycles for the curing of composites

using embedded resistive heating elements. Heating

and cooling was controlled internal to the mater-

ials rather than through external heating and cooling

equipment.

To get to a better understanding of the cur-

ing process Thomas et al. performed experimental

characterizations of autoclave-cured glass–epoxy com-

posite laminates at Washington University in St. Louis,

MO [28.15]. From the results of over 100 experimental

autoclave curing runs in 1996, the study team sought

to verify shrinking horizon model predictive control

(SHMPC) to predict and control the thickness and void

content of their composite materials. At the Korea Ad-

vanced Institute of Science and Technology, Kim and

Lee reported the results from their experiments with

composites to reduce fabrication thermal residue stress

of hybrid co-cured structures through the use of dielec-

trometry [28.16]. Ghasemi-Nejhad reported in 2005 his

method of manufacturing and testing of active com-

posite panels with embedded piezoelectric sensors and

actuators [28.17]. The embedded sensors and actuators

were used to minimize the problems associated with

layering composites such as voids and weak bonds.

Pantelelis et al. of the National Technical University of

Athens presented a computer simulation tool in 2004

that was coupled with a numerical optimization method

developed for use in the optimal design of cure cy-

cles in the production of thermoset-matrix composite

parts [28.18]. The presenters, Georg et al. from Boe-

ing, Madsen and Teng from Northrup Grumman, and

Courdji from Convergent Manufacturing, discussed the

exploration of composites processing and producibility

by analysis [28.19]. The discussion centered predomi-

nately on accelerated insertion of materials-composites

(AIM-C) processing and producibility modules and the

results thatcould beachieve bylinking itwith the robust

design computation system (RDCS). AIM-C was being

conducted jointly by the Naval Air Systems Command

(NASC) and the Defense Advance Research Projects

Agency (DARPA). In particular, AIM-C sought to sig-

niﬁcantly reduce the time and cost of inserting new

materials.

For more than 10 years, the University of Delaware

has been formulating advanced curing controls in its

Department of Chemical Engineering and Center for

Composite Materials.A team consisting of Pillai, Beris,

and Dhurjati developed an expert system toolin 1997 to

operate an autoclave for the intelligent curing of com-

posite materials [28.20]. At the same facility in 2002,

Michaud et al. developed a robust, simulation-based

optimization and control method methodology to iden-

tify and implement the optimal curing conditions for

thick-sectioned resin transfer molding (RTM) compos-

ite manufacturing [28.21].

To get better validation results for the curing pro-

cess, Arms et al. developed a technique for remote

powering and bidirectional communication with a net-

work of micro-miniature, multichannel addressable

sensing modules [28.22]. Use of the embedded sensors

sought to eliminate the damage that is commonly in-

curred when removing thermocouples from composite

materials in the post-curing phase of production. Urya-

sev and Trindade also studied methods to achieve better

validation results and settled on a method combining an

Part C 28.1

Distributed Agent Software for Automation 28.2 Industrial Agent Architecture 473

analytical model and experimental test data for optimal

determination of failure tolerance limits [28.23].

Alternative curing schemes are also being devel-

oped. Electron beam (EB) curing is a technique that

researchers at Acsion in Manitoba, Canada say could

have a massive impact in the aerospace and defense

industries, especially for the composite materials used

in aircraft wings [28.24]. In electron beam curing, the

incoming high-speed electrons knock off other elec-

trons in the polymer resin, causing the material to cure.

X-rays generatedfrom the high-energy beams canpene-

trate even deeper into the composite, yielding a uniform

cure in material several centimeters thick. The process

takes only minutes compared with hours for conven-

tional autoclave curing. Eight independent studies have

shown potential manufacturing cost savings of 26–65%

for prototyping alone. This could rise to as much as

90–95% [28.24]. The National Aeronautics and Space

Administration (NASA) encouraged the advancement

of the EB curing process through a small business

innovation research (SBIR) grant awarded to Science

Research Laboratory (SRL), Inc. As a result of the work

conducted, a new curingmethod was producedthat used

automated tape placement with the electron beam. The

new method is known as in situ electron beam curing

and the SRL-devised system is being utilized at the

NASA Marshall Space Flight Center [28.25].

Commercial interests have been developing other

methods and products to supplement the curing con-

trols that have already been put into use in the factory.

The Blair Rubber Company of Seville, OH has devel-

oped very precise process speciﬁcations for vulcanizing

rubber products in curing autoclaves. Multichannel

communications modules are in common use with

thermocouples. In particular, Data Translation, Inc.

has universal serial bus (USB) thermocouple measure-

ment modules that can monitor several thermocouple

devices simultaneously. Flexible thermocouple man-

agement programs have been developed to minimize

the costs associated with thermocouple wear and tear.

Thermocouples are reworkedand/or recalibratedand re-

turned to service through this type of program. Other

thermocouple management programs are involved with

manual and automatic device slaving, intelligent ther-

mocouples that protect against ground fault current,

automatic compensation, and automatic swapping of

lead thermocouples [28.26–30].

Even the design and manufacture of thermocou-

ples has become specialized. Customized wiring is

provided within thermocouples by manufacturers de-

pending upon the industrial application and/or the

environment in which the sensors will be placed. Within

autoclave-based composite applications, there are at

least seven types of thermocouples.

28.2 Industrial Agent Architecture

Industrial agents have been designed to execute on pro-

grammable logic controllers (PLCs). The agents are

built on workstations, compiled, and then downloaded

to the PLCs [28.31–33]. The downloading of the agents

is directed by central management software that exe-

cutes an agent assignment script. A decision was made

to expand the PLC ﬁrmware to enable for the program-

ming of advanced decision-making engines.

The agents have three main parts:

1. Reasoning

2. Data table interface

3. Execution control.

The ﬁrst part is the actual brain of the agent.

The brain is composed of behaviors, which execute as

needed, as prescribed in the process operations. Dy-

namically generated events trigger the execution of

behaviors in the agent to follow a particular sequence of

procedures. An industrial agent is essentially reactive,

but its software infrastructure permits the incorporation

of proactivebehaviors.The datatable interface serves as

the repository of control and agent data. The reasoning

part sets values in the data table to estimable the control

loops. The control part is the control execution level.

This part is in charge of executing control loops in a pe-

riodic or continuous fashion to maintain steady-state

conditions during the execution of the operations.

As shown in Fig. 28.2, a hierarchical library of

interrelated agents represents the composite curing ap-

plication. Each node of the library corresponds to an

agent of a given type such as a thermocouple or auto-

clave. Each agent will expose a set of user-conﬁgurable

attributes to give it a personality and operational ranges.

The interagent communication is based on the FIPA

(Foundation for Intelligent Physical Agents) [28.34]

language speciﬁcation. The content of the agent

message is written in a job description language

Part C 28.2

474 Part C Automation Design: Theory, Elements, and Methods

Hierarchical

structure

Agent

Attributes

Reasoning

/Java

Interface

IEC-61131

Talks IEEE-FIPA over

Ethernet/IP and ControlNet

Data table holds I/0

symbols

Talks common information

protocol (CIP)

PLCs

Composite curing system

Curing

Autoclave01

Cooler

Heater

TC07...

TC08...

TC09...

TC10...

TC11...

TC12...

TC13...

TC14...

TC15...

Fig. 28.2 Industrial agent components. CIP – common information protocol

(JDL) [28.31–33]. The industrial platform is based on

Rockwell Automation’s PLCs and NetLinx networking.

28.2.1 Agent Design

and Partitioning Perspective

The beneﬁts of using agents can be observed in three

main technological aspects:

1. System design

2. Upgrading

3. Runtime.

For an appropriate system design technique, the ef-

fects of changing components (adding, removing, and

altering) must be measured. The agent technology pro-

vides new techniques todesign truly distributed systems

by using libraries of functionality where the designers

specify the components and their corresponding be-

haviors in generic terms. The libraries of functionality

augment the system design capabilities by eliminating

the need to redesign the whole system when changes

are made.

System upgradability is traditionally a difﬁcult task

in industrial systems due to the hard-coupled dependen-

cies among the controlled components (tightly coupled

systems). In a nonagent system, the logical dependen-

cies among the components must be hard-coded into

the software during design time. Therefore, the ef-

fect of changing one component cascades throughout

the system, forcing the modiﬁcation of other com-

ponents, and so on. In an agent-based system, one

attempts to eliminate this cascading effect by emphasiz-

ing loosely coupled relationships among the controlled

components. Agents generatedynamic interconnections

with counterparts via agent messages. The dynamically

emerging interconnectivity effects the agents’ behaviors

by locally changing their world view.

The runtime perspective of agent-based systems is

helped by the agent’s capability (or service) of in-

frastructure discovery. Using the dynamic discovery

infrastructure, an agent can initiate the discovery of

other agents than can help in responding to a par-

ticular event. As opposed to having a speciﬁc set of

predeﬁned plans or agents to talk to under particular cir-

cumstances, the agents opt to discover their associates

that can currently supply a solution. This distributed

nature of agent technology opens the door to creating

a more survivable system by eliminating single points

of failure.

A very signiﬁcant aspect of the agent-based control

technology is system scalability. An agent solution can

be scaled up and down to ﬁt different system sizes.

A powerful feature of agents is the ability for

interagent communications to detect, isolate, and ac-

commodate component failures. In the case of a mul-

tithermocouple system, agent-to-agent communication

is directed at validating normative temperature read-

ings. Any variations from the group’s tendency are

Part C 28.2